From the course: AI Trends

Unlock this course with a free trial

Join today to access over 25,300 courses taught by industry experts.

Gemini Diffusion model

Gemini Diffusion model

From the course: AI Trends

Start my 1-month free trial Buy for my team

Gemini Diffusion model

“

- Most AI models, guess the next word, but Google Gemini Diffusion does something different, it edits noise into meaning, and that could be the next big thing. Now, the thing that's interesting about Gemini is that it flips how language models work. Instead of predicting words one by one, it refines noise to generate entire sections of text at once, and that means faster, more coherent responses. In fact, it clocks in at close to 1500 tokens per second. If you take a look at how fast some of the most popular models work, you can see that the fastest Gemini 2.5 Flash is right now running at 331 tokens per second. So 1500 tokens is lightning fast. To show you the difference in speed, I'm going to run one of these sample prompts in Diffusion. Then I'll use 2.5 Pro to compare it as well as ChatGPT-4o. This particular prompt is going to require some translation and that means I can't use Gemini Flash, which would be a more…

Contents